Clostridium difficile produces an assortment of gastrointestinal diseases in humans and animals ranging from mild diarrhea to life threatening pseudomembranous colitis. It is widely accepted that C. difficile causes pseudomembranous colitis in humans as a result of the elimination of the normal flora of the colon by antibiotic usage and subsequent growth of this toxin producing bacterium. The disease usually occurs in hospitalized patients where it causes a massive diarrhea with extensive inflammation of the colon. Mortality rates as high as 44% have been reported. Treatment of the disease is possible but relies on a proper diagnosis which may be accomplished by establishing the presence of the causation organism and demonstrating the characteristic lesions in the colon.
C. difficile produces two toxins, designated Toxin A and Toxin B, that are cytotoxic for tissue cultured mammalian cells. In addition to its cytotoxicity, Toxin A also possesses enterotoxin activity. Toxin A causes an accumulation of fluid in the intestine loops and causes extensive damage to the gut mucosa. Although both toxins are typically produced during disease, the principle laboratory diagnostic methods for detection of C. difficile have been directed towards detection of Toxin A. For example, a commercial latex test for the presence of Toxin A was marketed by Marion Scientific, a division of Marion Laboratories, Inc. of Kansas City, Mo. This commercial test was, unfortunately, found by subsequent researchers to be non-specific for Toxin A. See Lyerly et al., J. Clin. Micro, 23: 622-623 (1986).
One method for detecting pathogenic C. difficile involves the culture of human feces, which requires specialized facilities for long periods of incubation and which has the disadvantage of interference by non-pathogenic C. difficile strains. Another method for detecting Toxin B, but which does not work well for detecting Toxin A, is cytotoxicity assays using tissue culture cells. Because Toxin A is a much less potent cytotoxin it is more difficult to detect in tissue culture assays.
Other methods developed for Toxin A detection have been based on the use of specific antibodies to Toxin A. These methods include the enzyme-linked immunosorbent assay (ELISA) as taught by Lyerly et al., J. Clin. Micro., 17: 72-78 (1983) and Laughton et al., J. Infect. Dis., 149: 781-788 (1984). Lyerly et al. reported detection of Toxin A at 1 ng (5 ng/ml) quantities while Laughton et al. reported detection at 0.1 ng (1.0 ng/ml) levels. An ELISA assay using a monoclonal antibody to Toxin A was reported by Lyerly et al., J. Clin. Micro., 21: 12-14 (1985), which could detect 4 ng (0.02 .mu.g/ml) of Toxin A. This detection level is usually adequate to detect the toxin in stool samples from patients with C. difficile-related diarrhea. Another antibody dependent test is the Latex Agglutination Test (LAT) wherein antibody is immobilized on latex beads and agglutination of said beads by soluble Toxin A is visualized.
U.S. Pat. No. 4,530,833 to Wilkins et al. teaches a method of purifying Toxin A from cell culture supernatant which comprises an ion-exchange chromatography step followed by isoelectric precipitation. The patent teaches that purified Toxin A is achieved by adjusting the pH and molarity of a solution of impure Toxin A to, preferably, 5.5 and 0.01, respectively. One disadvantage of using isoelectric precipitation for protein purification is that other proteins with similar properties as the desired protein aggregate and coprecipitate with the desired protein to form the isoelectric precipitate. Thus, the precipitate must be washed by resuspension and recentrifugation. This results in loss of the desired protein. In addition, the process prior to and including the precipitation step must be very carefully controlled to control the precipitation step. If the fractions from the ion-exchange column are not carefully and consistently pooled, more or less proteins will co-precipitate with the Toxin A.
Another disadvantage of isoelectric precipitation with respect to purification of Toxin A is that denaturation of Toxin A may result if high speeds are used to pellet the precipitated protein. Thus, the protein will not be active in immunoassays or when used as an immunogen for antibody production. Isoelectric precipitation of proteins may also cause changes in the tertiary structure of the protein. In particular, Toxin A is an extremely hydrophobic protein and its exposure to a low ionic strength buffer for several wash steps can result in its partial denaturation.
Another method of Toxin A purification is taught in U.S. Pat. No. 5,098,826 to Wilkins et al. This method comprises contacting an impure solution of Toxin A with an immobilized reagent containing one or more of the terminal non-reducing structures characteristic of any of the human antigens X, Y, or I. These structures are known to be polysaccharides, and they can be immobilized by binding them to an insoluble matrix, e.g., silica gel, agarose, latex beads, or the like. The Toxin A binds to the immobilized reagent and is then eluted from the reagent in a more pure form. This method requires one of the antigens X, Y, or I, which are relatively expensive to obtain, and thus is an impractical method for purification of useful quantities of Toxin A.
Production of polyclonal antibodies to Toxin A is known in the art. For example, Ehrich et al., Inf. Immun. 28: 1041-43 (1980) teaches a procedure of culturing C. difficile and inoculating rabbits to produce polyclonal antibodies to an impure C. difficile preparation. Such polyclonal antibodies are not mono-specific for toxin because of the large number of proteins present in culture supernatant. Thus, the performance of toxin specific polyclonal based EIA have been disappointing because of the relatively low toxin specific titer and the high level of nonspecific reactivity of the polyclonal antisera.
Therefore there is a need for a method for the purification of Toxin A as well as a rapid and accurate assay for diagnosis and detection of C. difficile Toxin A.
It can be seen by the above that a method for purifying Toxin A which results in a consistently pure and non-denatured protein and which can be used to purify large quantities of protein is needed. Antibodies produced by such a highly purified, active protein will themselves be of higher activity.